arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.04217cs.LGcs.CVstat.ML
收藏

深度学习层的谱张量训练参数化

Spectral Tensor Train Parameterization of Deep Learning Layers

ETH Zurich(苏黎世联邦理工学院);HSE University(俄罗斯国立高等经济大学);KU Leuven(鲁汶大学)

Anton Obukhov, Maxim Rakhuba, Alexander Liniger, Zhiwu Huang, Stamatios Georgoulis, Dengxin Dai, Luc Van Gool

摘要:

我们研究了在深度学习背景下嵌入谱属性的权重矩阵的低秩参数化。低秩属性导致参数效率,并允许在计算映射时采取计算捷径。在优化问题中,谱特性常常受到约束,从而导致更好的模型和优化的稳定性。首先,我们看看紧凑的SVD参数化的权重矩阵,并确定冗余源的参数化。我们进一步应用张量训练(TT)分解的紧凑SVD组件,并提出了一个非冗余的可微参数化固定TT秩张量流形,称为谱张量训练参数化(STTP)。我们展示了神经网络压缩在图像分类设置中的效果,以及在生成对抗训练设置中的压缩和改进的训练稳定性。

英文摘要:

We study low-rank parameterizations of weight matrices with embedded spectral properties in the Deep Learning context. The low-rank property leads to parameter efficiency and permits taking computational shortcuts when computing mappings. Spectral properties are often subject to constraints in optimization problems, leading to better models and stability of optimization. We start by looking at the compact SVD parameterization of weight matrices and identifying redundancy sources in the parameterization. We further apply the Tensor Train (TT) decomposition to the compact SVD components, and propose a non-redundant differentiable parameterization of fixed TT-rank tensor manifolds, termed the Spectral Tensor Train Parameterization (STTP). We demonstrate the effects of neural network compression in the image classification setting and both compression and improved training stability in the generative adversarial training setting.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑